If the rot starts at the core of your education curriculum there is no saving your dependence on Microsoft.
I always found this choice puzzling to teach people proprietary technologies in a public institution. This was before DotNet core and VSCode was a thing and Microsoft hadnt whitewashed themselves to look like an open source friendly brand yet.
In my experience, getting the periphery stuff is harder than the contents of the course a good chunk of the time for beginners.
TBF, the curriculum being MS based can mean very little if the concepts taught in are valuable enough. I've briefly looked at the project linked in your user description, and they don't look nice and absolutely not tinted by MS influence.
It is indeed dancing with the devil, but if MS forks the money to renew the whole university's computer park, clear all the licensing issues and train part of the staff, it can be a boon for the university.
My uni had a deal with Sun (RIP), many basic courses were in Java, all our system programming course we're Solaris targeted, all servers were Solaris anyway so our code had to run there. It's a pretty basic arrangement IMHO.
More importantly, using Microsoft at scale can leave your organization fundamentally insecure. The obscure, insecure defaults are, at best, dangerous missteps and, at worst, borderline negligent. I’m convinced that only a small fraction of enterprises using Microsoft have the expertise and budget required to secure it properly.
My personal view is that if your organization depends heavily on Microsoft, it’s not serious about security, whether they’re aware of it or not.
They’re just hedging that American politics will stop licking the car battery.
While I applaud the use of alternatives to windows and it's apps, universities teach it because it is what their graduates will use in the real world. Governments use it because while it has it's flaws, it mostly works and is a universal standard. It's the toyota of operating systems. The parts and manpower to repair it and use it are available everywhere, and it's cheap and reliable.
And the thing is 99.99% of the time everything works just fine. I think these governments often struggle with moving off of them because they find that making the common case worse is not a trade off that most of their users want.
The path taken by Blender(propreiety initially to open source) to reach industry lead would to me seem the most viable to make a dent.
In that i think best cost effective options like WPSOffice or Corel Suite , would be a good option.They have the professional usability in the interface and functionality.
Corel is basically leaving the market wide , by mostly collecting rent from lawfirms as they are well taken care of there.Considering they used to have viable Linux options , seems a lack of vision theer to pick up marketshare.
I feel if the TUs were required to dogfood this, especially if generously funded such that startups could come along and provide the same service and support, that it could be a great positive externality
It is very distressing how many organizations have become dependant on Microsoft and the US cloud for core services. I hope that an unintended consequence of the current US administration's approach is that this becomes less so.
I was one of the people fighting for keeping Unix when the UU went to Exchange. It was a drama: instable af, the MS consultants could not keep it running even for 24 hours at a time while unix had 0 issues and kept chugging along (I don't remember what Unix: I think it was SunOS/Solaris). It was forced through at great cost and effort but of course sponsored by MS. It sucked for years to come.
I was at the UvA too when they moved to, equally instable MS stuff too: I worked behind some of the last Sun machines and got to take a palet of sparcstations, ultras and an e450 home when they got phased out (I still have them and they are still working, of course). Could have all been Linux now but MS was so aggressive and no one listened to profs or students, even in all tech deps who were all vehemently against the move.
Its' an awful sight. What's worse is that there's no argument for this extra cost (apart from maybe vendor lock-in), and now no one knows who to blame for the big bill that comes in every month.
By injecting CoPilot into it without customer validation is going to be very costly.
step 2. cost savings by firing them all
step 3. we get locked in
step 4. oh no how did this happen
Actual independence would require a great deal of competence, expenditure, hard work, long-range planning, and time living unhappily far from any optimum.
While the Dutch obviously know how to do that - nobody in America is keeping the North Sea at bay for them - I would not bet that they'll actually do it here.
(If the EU never figures out how to allow nimble startups companies to exist, then forget about it)
Today, a desktop is the primary epistemological tool; only a FLOSS one is suitable for study, as it reveals what it does and is, or at least can be, shaped according to need. Most are simply IT illiterate, like professors from the 1950s unable to read and write fluently, except they still haven't realised it...
Microsoft's partnership/JV with 21Vianet in China shows that it can be done (with limitations on features) and this is probably the right direction to go.
FOSS to the rescue:
* LibreOffice, instead of MS Office.
* Thunderbird, instead of Outlook (even though I don't like the direction Thunderbird has been going).
* NextCloud, instead of one drive (and there are other alternatives, more FOSS-friendly or less so)
* Matrix/IRC client plus Jitsi, instead of Teams.
and they will do just fine - on Windows or on Linux.
"It seems to me like all the things you're doing can and should be automated at a larger scale."
Ten years ago when I recalled this I felt sort of good about the prediction. What I predicted pretty much happened.
That sysadmin guy has become some sort of CIO and seems to be doing well.
I did not anticipate the loss of data sovereignty.
.... and now I'm doing like 50% SRE/devops. Who's the sysadmin now, but without physical control of our data?
I was kind of hoping that the GDPR would be that kick, as it's clear to anyone that Microsoft, or really any other major US corporation, can't actually satisfy that completely as long as they have their tendrils in its European subsidiaries and that the US can compel them give up the information they care about. But this is a rather large elephant the EU has elected to ignore since that's the path of least resistance.
The EU actually realising the using Microsoft as your foundation does break the GDPR and fining the relevant institutions (and fixing itself on that same front!), or relevant institutions being embargoed to the point of not being able to use Microsoft products, as is apparently now the case with the ICC, will probably kick Europe into gear, but it needs to be a solid kick, and not just an institution here and there.
I hate that people are incapable of using Libreoffice and mailing documents around, but modern users are addicted to "the cloud", and it's my understanding there's no EU centric competitor to those two giants.